ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.51 by root, Sat Jun 24 19:14:04 2006 UTC vs.
Revision 1.183 by root, Sun Sep 12 03:40:05 2010 UTC

…		…
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	use IO::AIO;	7	use IO::AIO;
8		8
9	aio_open "/etc/passwd", O_RDONLY, 0, sub {	9	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
10	my ($fh) = @_;	10	my $fh = shift
		11	or die "/etc/passwd: $!";
11	...	12	...
12	};	13	};
13		14
14	aio_unlink "/tmp/file", sub { };	15	aio_unlink "/tmp/file", sub { };
15		16
16	aio_read $fh, 30000, 1024, $buffer, 0, sub {	17	aio_read $fh, 30000, 1024, $buffer, 0, sub {
17	$_[0] > 0 or die "read error: $!";	18	$_[0] > 0 or die "read error: $!";
18	};	19	};
19		20
20	# AnyEvent	21	# version 2+ has request and group objects
21	open my $fh, "<&=" . IO::AIO::poll_fileno or die "$!";	22	use IO::AIO 2;
22	my $w = AnyEvent->io (fh => $fh, poll => 'r', cb => sub { IO::AIO::poll_cb });
23		23
24	# Event	24	aioreq_pri 4; # give next request a very high priority
25	Event->io (fd => IO::AIO::poll_fileno,	25	my $req = aio_unlink "/tmp/file", sub { };
26	poll => 'r',	26	$req->cancel; # cancel request if still in queue
27	cb => \&IO::AIO::poll_cb);
28		27
29	# Glib/Gtk2	28	my $grp = aio_group sub { print "all stats done\n" };
30	add_watch Glib::IO IO::AIO::poll_fileno,	29	add $grp aio_stat "..." for ...;
31	in => sub { IO::AIO::poll_cb; 1 };
32
33	# Tk
34	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
35	readable => \&IO::AIO::poll_cb);
36
37	# Danga::Socket
38	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
39	\&IO::AIO::poll_cb);
40
41		30
42	=head1 DESCRIPTION	31	=head1 DESCRIPTION
43		32
44	This module implements asynchronous I/O using whatever means your	33	This module implements asynchronous I/O using whatever means your
45	operating system supports.	34	operating system supports. It is implemented as an interface to C<libeio>
		35	(L<http://software.schmorp.de/pkg/libeio.html>).
46		36
		37	Asynchronous means that operations that can normally block your program
		38	(e.g. reading from disk) will be done asynchronously: the operation
		39	will still block, but you can do something else in the meantime. This
		40	is extremely useful for programs that need to stay interactive even
		41	when doing heavy I/O (GUI programs, high performance network servers
		42	etc.), but can also be used to easily do operations in parallel that are
		43	normally done sequentially, e.g. stat'ing many files, which is much faster
		44	on a RAID volume or over NFS when you do a number of stat operations
		45	concurrently.
		46
		47	While most of this works on all types of file descriptors (for
		48	example sockets), using these functions on file descriptors that
		49	support nonblocking operation (again, sockets, pipes etc.) is
		50	very inefficient. Use an event loop for that (such as the L<EV>
		51	module): IO::AIO will naturally fit into such an event loop itself.
		52
47	Currently, a number of threads are started that execute your read/writes	53	In this version, a number of threads are started that execute your
48	and signal their completion. You don't need thread support in your libc or	54	requests and signal their completion. You don't need thread support
49	perl, and the threads created by this module will not be visible to the	55	in perl, and the threads created by this module will not be visible
50	pthreads library. In the future, this module might make use of the native	56	to perl. In the future, this module might make use of the native aio
51	aio functions available on many operating systems. However, they are often	57	functions available on many operating systems. However, they are often
52	not well-supported (Linux doesn't allow them on normal files currently,	58	not well-supported or restricted (GNU/Linux doesn't allow them on normal
53	for example), and they would only support aio_read and aio_write, so the	59	files currently, for example), and they would only support aio_read and
54	remaining functionality would have to be implemented using threads anyway.	60	aio_write, so the remaining functionality would have to be implemented
		61	using threads anyway.
55		62
56	Although the module will work with in the presence of other threads, it is	63	Although the module will work in the presence of other (Perl-) threads,
57	currently not reentrant, so use appropriate locking yourself, always call	64	it is currently not reentrant in any way, so use appropriate locking
58	C<poll_cb> from within the same thread, or never call C<poll_cb> (or other	65	yourself, always call C<poll_cb> from within the same thread, or never
59	C<aio_> functions) recursively.	66	call C<poll_cb> (or other C<aio_> functions) recursively.
		67
		68	=head2 EXAMPLE
		69
		70	This is a simple example that uses the EV module and loads
		71	F</etc/passwd> asynchronously:
		72
		73	use Fcntl;
		74	use EV;
		75	use IO::AIO;
		76
		77	# register the IO::AIO callback with EV
		78	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		79
		80	# queue the request to open /etc/passwd
		81	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
		82	my $fh = shift
		83	or die "error while opening: $!";
		84
		85	# stat'ing filehandles is generally non-blocking
		86	my $size = -s $fh;
		87
		88	# queue a request to read the file
		89	my $contents;
		90	aio_read $fh, 0, $size, $contents, 0, sub {
		91	$_[0] == $size
		92	or die "short read: $!";
		93
		94	close $fh;
		95
		96	# file contents now in $contents
		97	print $contents;
		98
		99	# exit event loop and program
		100	EV::unloop;
		101	};
		102	};
		103
		104	# possibly queue up other requests, or open GUI windows,
		105	# check for sockets etc. etc.
		106
		107	# process events as long as there are some:
		108	EV::loop;
		109
		110	=head1 REQUEST ANATOMY AND LIFETIME
		111
		112	Every C<aio_*> function creates a request. which is a C data structure not
		113	directly visible to Perl.
		114
		115	If called in non-void context, every request function returns a Perl
		116	object representing the request. In void context, nothing is returned,
		117	which saves a bit of memory.
		118
		119	The perl object is a fairly standard ref-to-hash object. The hash contents
		120	are not used by IO::AIO so you are free to store anything you like in it.
		121
		122	During their existance, aio requests travel through the following states,
		123	in order:
		124
		125	=over 4
		126
		127	=item ready
		128
		129	Immediately after a request is created it is put into the ready state,
		130	waiting for a thread to execute it.
		131
		132	=item execute
		133
		134	A thread has accepted the request for processing and is currently
		135	executing it (e.g. blocking in read).
		136
		137	=item pending
		138
		139	The request has been executed and is waiting for result processing.
		140
		141	While request submission and execution is fully asynchronous, result
		142	processing is not and relies on the perl interpreter calling C<poll_cb>
		143	(or another function with the same effect).
		144
		145	=item result
		146
		147	The request results are processed synchronously by C<poll_cb>.
		148
		149	The C<poll_cb> function will process all outstanding aio requests by
		150	calling their callbacks, freeing memory associated with them and managing
		151	any groups they are contained in.
		152
		153	=item done
		154
		155	Request has reached the end of its lifetime and holds no resources anymore
		156	(except possibly for the Perl object, but its connection to the actual
		157	aio request is severed and calling its methods will either do nothing or
		158	result in a runtime error).
		159
		160	=back
60		161
61	=cut	162	=cut
62		163
63	package IO::AIO;	164	package IO::AIO;
64		165
65	no warnings;	166	use Carp ();
66	use strict 'vars';	167
		168	use common::sense;
67		169
68	use base 'Exporter';	170	use base 'Exporter';
69		171
70	BEGIN {	172	BEGIN {
71	our $VERSION = '1.8';	173	our $VERSION = '3.65';
72		174
73	our @EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat	175	our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close
74	aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink	176	aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
75	aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move);	177	aio_scandir aio_symlink aio_readlink aio_sync aio_fsync
76	our @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs);	178	aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead
		179	aio_rename aio_link aio_move aio_copy aio_group
		180	aio_nop aio_mknod aio_load aio_rmtree aio_mkdir aio_chown
		181	aio_chmod aio_utime aio_truncate
		182	aio_msync aio_mtouch aio_mlock aio_mlockall
		183	aio_statvfs);
		184
		185	our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice));
		186	our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush
		187	min_parallel max_parallel max_idle
		188	nreqs nready npending nthreads
		189	max_poll_time max_poll_reqs
		190	sendfile fadvise madvise
		191	mmap munmap munlock munlockall);
		192
		193	push @AIO_REQ, qw(aio_busy); # not exported
		194
		195	@IO::AIO::GRP::ISA = 'IO::AIO::REQ';
77		196
78	require XSLoader;	197	require XSLoader;
79	XSLoader::load ("IO::AIO", $VERSION);	198	XSLoader::load ("IO::AIO", $VERSION);
80	}	199	}
81		200
82	=head1 FUNCTIONS	201	=head1 FUNCTIONS
83		202
		203	=head2 QUICK OVERVIEW
		204
		205	This section simply lists the prototypes of the most important functions
		206	for quick reference. See the following sections for function-by-function
		207	documentation.
		208
		209	aio_open $pathname, $flags, $mode, $callback->($fh)
		210	aio_close $fh, $callback->($status)
		211	aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		212	aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
		213	aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
		214	aio_readahead $fh,$offset,$length, $callback->($retval)
		215	aio_stat $fh_or_path, $callback->($status)
		216	aio_lstat $fh, $callback->($status)
		217	aio_statvfs $fh_or_path, $callback->($statvfs)
		218	aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		219	aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		220	aio_truncate $fh_or_path, $offset, $callback->($status)
		221	aio_chmod $fh_or_path, $mode, $callback->($status)
		222	aio_unlink $pathname, $callback->($status)
		223	aio_mknod $path, $mode, $dev, $callback->($status)
		224	aio_link $srcpath, $dstpath, $callback->($status)
		225	aio_symlink $srcpath, $dstpath, $callback->($status)
		226	aio_readlink $path, $callback->($link)
		227	aio_rename $srcpath, $dstpath, $callback->($status)
		228	aio_mkdir $pathname, $mode, $callback->($status)
		229	aio_rmdir $pathname, $callback->($status)
		230	aio_readdir $pathname, $callback->($entries)
		231	aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		232	IO::AIO::READDIR_DENTS IO::AIO::READDIR_DIRS_FIRST
		233	IO::AIO::READDIR_STAT_ORDER IO::AIO::READDIR_FOUND_UNKNOWN
		234	aio_load $path, $data, $callback->($status)
		235	aio_copy $srcpath, $dstpath, $callback->($status)
		236	aio_move $srcpath, $dstpath, $callback->($status)
		237	aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
		238	aio_rmtree $path, $callback->($status)
		239	aio_sync $callback->($status)
		240	aio_fsync $fh, $callback->($status)
		241	aio_fdatasync $fh, $callback->($status)
		242	aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		243	aio_pathsync $path, $callback->($status)
		244	aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		245	aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		246	aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		247	aio_mlockall $flags, $callback->($status)
		248	aio_group $callback->(...)
		249	aio_nop $callback->()
		250
		251	$prev_pri = aioreq_pri [$pri]
		252	aioreq_nice $pri_adjust
		253
		254	IO::AIO::poll_wait
		255	IO::AIO::poll_cb
		256	IO::AIO::poll
		257	IO::AIO::flush
		258	IO::AIO::max_poll_reqs $nreqs
		259	IO::AIO::max_poll_time $seconds
		260	IO::AIO::min_parallel $nthreads
		261	IO::AIO::max_parallel $nthreads
		262	IO::AIO::max_idle $nthreads
		263	IO::AIO::max_outstanding $maxreqs
		264	IO::AIO::nreqs
		265	IO::AIO::nready
		266	IO::AIO::npending
		267
		268	IO::AIO::sendfile $ofh, $ifh, $offset, $count
		269	IO::AIO::fadvise $fh, $offset, $len, $advice
		270	IO::AIO::munlock $scalar, $offset = 0, $length = undef
		271	IO::AIO::munlockall
		272
84	=head2 AIO FUNCTIONS	273	=head2 AIO REQUEST FUNCTIONS
85		274
86	All the C<aio_*> calls are more or less thin wrappers around the syscall	275	All the C<aio_*> calls are more or less thin wrappers around the syscall
87	with the same name (sans C<aio_>). The arguments are similar or identical,	276	with the same name (sans C<aio_>). The arguments are similar or identical,
88	and they all accept an additional (and optional) C<$callback> argument	277	and they all accept an additional (and optional) C<$callback> argument
89	which must be a code reference. This code reference will get called with	278	which must be a code reference. This code reference will get called with
90	the syscall return code (e.g. most syscalls return C<-1> on error, unlike	279	the syscall return code (e.g. most syscalls return C<-1> on error, unlike
91	perl, which usually delivers "false") as it's sole argument when the given	280	perl, which usually delivers "false") as its sole argument after the given
92	syscall has been executed asynchronously.	281	syscall has been executed asynchronously.
93		282
94	All functions expecting a filehandle keep a copy of the filehandle	283	All functions expecting a filehandle keep a copy of the filehandle
95	internally until the request has finished.	284	internally until the request has finished.
96		285
		286	All functions return request objects of type L<IO::AIO::REQ> that allow
		287	further manipulation of those requests while they are in-flight.
		288
97	The pathnames you pass to these routines I<must> be absolute and	289	The pathnames you pass to these routines I<must> be absolute and
98	encoded in byte form. The reason for the former is that at the time the	290	encoded as octets. The reason for the former is that at the time the
99	request is being executed, the current working directory could have	291	request is being executed, the current working directory could have
100	changed. Alternatively, you can make sure that you never change the	292	changed. Alternatively, you can make sure that you never change the
101	current working directory.	293	current working directory anywhere in the program and then use relative
		294	paths.
102		295
103	To encode pathnames to byte form, either make sure you either: a)	296	To encode pathnames as octets, either make sure you either: a) always pass
104	always pass in filenames you got from outside (command line, readdir	297	in filenames you got from outside (command line, readdir etc.) without
105	etc.), b) are ASCII or ISO 8859-1, c) use the Encode module and encode	298	tinkering, b) are ASCII or ISO 8859-1, c) use the Encode module and encode
106	your pathnames to the locale (or other) encoding in effect in the user	299	your pathnames to the locale (or other) encoding in effect in the user
107	environment, d) use Glib::filename_from_unicode on unicode filenames or e)	300	environment, d) use Glib::filename_from_unicode on unicode filenames or e)
108	use something else.	301	use something else to ensure your scalar has the correct contents.
		302
		303	This works, btw. independent of the internal UTF-8 bit, which IO::AIO
		304	handles correctly whether it is set or not.
109		305
110	=over 4	306	=over 4
		307
		308	=item $prev_pri = aioreq_pri [$pri]
		309
		310	Returns the priority value that would be used for the next request and, if
		311	C<$pri> is given, sets the priority for the next aio request.
		312
		313	The default priority is C<0>, the minimum and maximum priorities are C<-4>
		314	and C<4>, respectively. Requests with higher priority will be serviced
		315	first.
		316
		317	The priority will be reset to C<0> after each call to one of the C<aio_*>
		318	functions.
		319
		320	Example: open a file with low priority, then read something from it with
		321	higher priority so the read request is serviced before other low priority
		322	open requests (potentially spamming the cache):
		323
		324	aioreq_pri -3;
		325	aio_open ..., sub {
		326	return unless $_[0];
		327
		328	aioreq_pri -2;
		329	aio_read $_[0], ..., sub {
		330	...
		331	};
		332	};
		333
		334
		335	=item aioreq_nice $pri_adjust
		336
		337	Similar to C<aioreq_pri>, but subtracts the given value from the current
		338	priority, so the effect is cumulative.
		339
111		340
112	=item aio_open $pathname, $flags, $mode, $callback->($fh)	341	=item aio_open $pathname, $flags, $mode, $callback->($fh)
113		342
114	Asynchronously open or create a file and call the callback with a newly	343	Asynchronously open or create a file and call the callback with a newly
115	created filehandle for the file.	344	created filehandle for the file.
…		…
121	list. They are the same as used by C<sysopen>.	350	list. They are the same as used by C<sysopen>.
122		351
123	Likewise, C<$mode> specifies the mode of the newly created file, if it	352	Likewise, C<$mode> specifies the mode of the newly created file, if it
124	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,	353	didn't exist and C<O_CREAT> has been given, just like perl's C<sysopen>,
125	except that it is mandatory (i.e. use C<0> if you don't create new files,	354	except that it is mandatory (i.e. use C<0> if you don't create new files,
126	and C<0666> or C<0777> if you do).	355	and C<0666> or C<0777> if you do). Note that the C<$mode> will be modified
		356	by the umask in effect then the request is being executed, so better never
		357	change the umask.
127		358
128	Example:	359	Example:
129		360
130	aio_open "/etc/passwd", O_RDONLY, 0, sub {	361	aio_open "/etc/passwd", IO::AIO::O_RDONLY, 0, sub {
131	if ($_[0]) {	362	if ($_[0]) {
132	print "open successful, fh is $_[0]\n";	363	print "open successful, fh is $_[0]\n";
133	...	364	...
134	} else {	365	} else {
135	die "open failed: $!\n";	366	die "open failed: $!\n";
136	}	367	}
137	};	368	};
138		369
		370
139	=item aio_close $fh, $callback->($status)	371	=item aio_close $fh, $callback->($status)
140		372
141	Asynchronously close a file and call the callback with the result	373	Asynchronously close a file and call the callback with the result
142	code. I<WARNING:> although accepted, you should not pass in a perl	374	code.
143	filehandle here, as perl will likely close the file descriptor another
144	time when the filehandle is destroyed. Normally, you can safely call perls
145	C<close> or just let filehandles go out of scope.
146		375
147	This is supposed to be a bug in the API, so that might change. It's	376	Unfortunately, you can't do this to perl. Perl I<insists> very strongly on
148	therefore best to avoid this function.	377	closing the file descriptor associated with the filehandle itself.
		378
		379	Therefore, C<aio_close> will not close the filehandle - instead it will
		380	use dup2 to overwrite the file descriptor with the write-end of a pipe
		381	(the pipe fd will be created on demand and will be cached).
		382
		383	Or in other words: the file descriptor will be closed, but it will not be
		384	free for reuse until the perl filehandle is closed.
		385
		386	=cut
149		387
150	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	388	=item aio_read $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
151		389
152	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)	390	=item aio_write $fh,$offset,$length, $data,$dataoffset, $callback->($retval)
153		391
154	Reads or writes C<length> bytes from the specified C<fh> and C<offset>	392	Reads or writes C<$length> bytes from or to the specified C<$fh> and
155	into the scalar given by C<data> and offset C<dataoffset> and calls the	393	C<$offset> into the scalar given by C<$data> and offset C<$dataoffset>
156	callback without the actual number of bytes read (or -1 on error, just	394	and calls the callback without the actual number of bytes read (or -1 on
157	like the syscall).	395	error, just like the syscall).
		396
		397	C<aio_read> will, like C<sysread>, shrink or grow the C<$data> scalar to
		398	offset plus the actual number of bytes read.
		399
		400	If C<$offset> is undefined, then the current file descriptor offset will
		401	be used (and updated), otherwise the file descriptor offset will not be
		402	changed by these calls.
		403
		404	If C<$length> is undefined in C<aio_write>, use the remaining length of
		405	C<$data>.
		406
		407	If C<$dataoffset> is less than zero, it will be counted from the end of
		408	C<$data>.
158		409
159	The C<$data> scalar I<MUST NOT> be modified in any way while the request	410	The C<$data> scalar I<MUST NOT> be modified in any way while the request
160	is outstanding. Modifying it can result in segfaults or WW3 (if the	411	is outstanding. Modifying it can result in segfaults or World War III (if
161	necessary/optional hardware is installed).	412	the necessary/optional hardware is installed).
162		413
163	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at	414	Example: Read 15 bytes at offset 7 into scalar C<$buffer>, starting at
164	offset C<0> within the scalar:	415	offset C<0> within the scalar:
165		416
166	aio_read $fh, 7, 15, $buffer, 0, sub {	417	aio_read $fh, 7, 15, $buffer, 0, sub {
167	$_[0] > 0 or die "read error: $!";	418	$_[0] > 0 or die "read error: $!";
168	print "read $_[0] bytes: <$buffer>\n";	419	print "read $_[0] bytes: <$buffer>\n";
169	};	420	};
170		421
171	=item aio_move $srcpath, $dstpath, $callback->($status)
172
173	[EXPERIMENTAL]
174
175	Try to move the I<file> (directories not supported as either source or destination)
176	from C<$srcpath> to C<$dstpath> and call the callback with the C<0> (error) or C<-1> ok.
177
178	This is a composite request that tries to rename(2) the file first. If
179	rename files with C<EXDEV>, it creates the destination file with mode 0200
180	and copies the contents of the source file into it using C<aio_sendfile>,
181	followed by restoring atime, mtime, access mode and uid/gid, in that
182	order, and unlinking the C<$srcpath>.
183
184	If an error occurs, the partial destination file will be unlinked, if
185	possible, except when setting atime, mtime, access mode and uid/gid, where
186	errors are being ignored.
187
188	=cut
189
190	sub aio_move($$$) {
191	my ($src, $dst, $cb) = @_;
192
193	aio_rename $src, $dst, sub {
194	if ($_[0] && $! == EXDEV) {
195	aio_open $src, O_RDONLY, 0, sub {
196	if (my $src_fh = $_[0]) {
197	my @stat = stat $src_fh;
198
199	aio_open $dst, O_WRONLY, 0200, sub {
200	if (my $dst_fh = $_[0]) {
201	aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
202	close $src_fh;
203
204	if ($_[0] == $stat[7]) {
205	utime $stat[8], $stat[9], $dst;
206	chmod $stat[2] & 07777, $dst_fh;
207	chown $stat[4], $stat[5], $dst_fh;
208	close $dst_fh;
209
210	aio_unlink $src, sub {
211	$cb->($_[0]);
212	};
213	} else {
214	my $errno = $!;
215	aio_unlink $dst, sub {
216	$! = $errno;
217	$cb->(-1);
218	};
219	}
220	};
221	} else {
222	$cb->(-1);
223	}
224	},
225
226	} else {
227	$cb->(-1);
228	}
229	};
230	} else {
231	$cb->($_[0]);
232	}
233	};
234	}
235		422
236	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)	423	=item aio_sendfile $out_fh, $in_fh, $in_offset, $length, $callback->($retval)
237		424
238	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts	425	Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts
239	reading at byte offset C<$in_offset>, and starts writing at the current	426	reading at byte offset C<$in_offset>, and starts writing at the current
…		…
241	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each	428	than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each
242	other.	429	other.
243		430
244	This call tries to make use of a native C<sendfile> syscall to provide	431	This call tries to make use of a native C<sendfile> syscall to provide
245	zero-copy operation. For this to work, C<$out_fh> should refer to a	432	zero-copy operation. For this to work, C<$out_fh> should refer to a
246	socket, and C<$in_fh> should refer to mmap'able file.	433	socket, and C<$in_fh> should refer to an mmap'able file.
247		434
248	If the native sendfile call fails or is not implemented, it will be	435	If a native sendfile cannot be found or it fails with C<ENOSYS>,
		436	C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or C<ENOTSOCK>,
249	emulated, so you can call C<aio_sendfile> on any type of filehandle	437	it will be emulated, so you can call C<aio_sendfile> on any type of
250	regardless of the limitations of the operating system.	438	filehandle regardless of the limitations of the operating system.
251		439
252	Please note, however, that C<aio_sendfile> can read more bytes from	440	Please note, however, that C<aio_sendfile> can read more bytes from
253	C<$in_fh> than are written, and there is no way to find out how many	441	C<$in_fh> than are written, and there is no way to find out how many
254	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only	442	bytes have been read from C<aio_sendfile> alone, as C<aio_sendfile> only
255	provides the number of bytes written to C<$out_fh>. Only if the result	443	provides the number of bytes written to C<$out_fh>. Only if the result
256	value equals C<$length> one can assume that C<$length> bytes have been	444	value equals C<$length> one can assume that C<$length> bytes have been
257	read.	445	read.
		446
258		447
259	=item aio_readahead $fh,$offset,$length, $callback->($retval)	448	=item aio_readahead $fh,$offset,$length, $callback->($retval)
260		449
261	C<aio_readahead> populates the page cache with data from a file so that	450	C<aio_readahead> populates the page cache with data from a file so that
262	subsequent reads from that file will not block on disk I/O. The C<$offset>	451	subsequent reads from that file will not block on disk I/O. The C<$offset>
…		…
268	file. The current file offset of the file is left unchanged.	457	file. The current file offset of the file is left unchanged.
269		458
270	If that syscall doesn't exist (likely if your OS isn't Linux) it will be	459	If that syscall doesn't exist (likely if your OS isn't Linux) it will be
271	emulated by simply reading the data, which would have a similar effect.	460	emulated by simply reading the data, which would have a similar effect.
272		461
		462
273	=item aio_stat $fh_or_path, $callback->($status)	463	=item aio_stat $fh_or_path, $callback->($status)
274		464
275	=item aio_lstat $fh, $callback->($status)	465	=item aio_lstat $fh, $callback->($status)
276		466
277	Works like perl's C<stat> or C<lstat> in void context. The callback will	467	Works like perl's C<stat> or C<lstat> in void context. The callback will
…		…
290	aio_stat "/etc/passwd", sub {	480	aio_stat "/etc/passwd", sub {
291	$_[0] and die "stat failed: $!";	481	$_[0] and die "stat failed: $!";
292	print "size is ", -s _, "\n";	482	print "size is ", -s _, "\n";
293	};	483	};
294		484
		485
		486	=item aio_statvfs $fh_or_path, $callback->($statvfs)
		487
		488	Works like the POSIX C<statvfs> or C<fstatvfs> syscalls, depending on
		489	whether a file handle or path was passed.
		490
		491	On success, the callback is passed a hash reference with the following
		492	members: C<bsize>, C<frsize>, C<blocks>, C<bfree>, C<bavail>, C<files>,
		493	C<ffree>, C<favail>, C<fsid>, C<flag> and C<namemax>. On failure, C<undef>
		494	is passed.
		495
		496	The following POSIX IO::AIO::ST_* constants are defined: C<ST_RDONLY> and
		497	C<ST_NOSUID>.
		498
		499	The following non-POSIX IO::AIO::ST_* flag masks are defined to
		500	their correct value when available, or to C<0> on systems that do
		501	not support them: C<ST_NODEV>, C<ST_NOEXEC>, C<ST_SYNCHRONOUS>,
		502	C<ST_MANDLOCK>, C<ST_WRITE>, C<ST_APPEND>, C<ST_IMMUTABLE>, C<ST_NOATIME>,
		503	C<ST_NODIRATIME> and C<ST_RELATIME>.
		504
		505	Example: stat C</wd> and dump out the data if successful.
		506
		507	aio_statvfs "/wd", sub {
		508	my $f = $_[0]
		509	or die "statvfs: $!";
		510
		511	use Data::Dumper;
		512	say Dumper $f;
		513	};
		514
		515	# result:
		516	{
		517	bsize => 1024,
		518	bfree => 4333064312,
		519	blocks => 10253828096,
		520	files => 2050765568,
		521	flag => 4096,
		522	favail => 2042092649,
		523	bavail => 4333064312,
		524	ffree => 2042092649,
		525	namemax => 255,
		526	frsize => 1024,
		527	fsid => 1810
		528	}
		529
		530
		531	=item aio_utime $fh_or_path, $atime, $mtime, $callback->($status)
		532
		533	Works like perl's C<utime> function (including the special case of $atime
		534	and $mtime being undef). Fractional times are supported if the underlying
		535	syscalls support them.
		536
		537	When called with a pathname, uses utimes(2) if available, otherwise
		538	utime(2). If called on a file descriptor, uses futimes(2) if available,
		539	otherwise returns ENOSYS, so this is not portable.
		540
		541	Examples:
		542
		543	# set atime and mtime to current time (basically touch(1)):
		544	aio_utime "path", undef, undef;
		545	# set atime to current time and mtime to beginning of the epoch:
		546	aio_utime "path", time, undef; # undef==0
		547
		548
		549	=item aio_chown $fh_or_path, $uid, $gid, $callback->($status)
		550
		551	Works like perl's C<chown> function, except that C<undef> for either $uid
		552	or $gid is being interpreted as "do not change" (but -1 can also be used).
		553
		554	Examples:
		555
		556	# same as "chown root path" in the shell:
		557	aio_chown "path", 0, -1;
		558	# same as above:
		559	aio_chown "path", 0, undef;
		560
		561
		562	=item aio_truncate $fh_or_path, $offset, $callback->($status)
		563
		564	Works like truncate(2) or ftruncate(2).
		565
		566
		567	=item aio_chmod $fh_or_path, $mode, $callback->($status)
		568
		569	Works like perl's C<chmod> function.
		570
		571
295	=item aio_unlink $pathname, $callback->($status)	572	=item aio_unlink $pathname, $callback->($status)
296		573
297	Asynchronously unlink (delete) a file and call the callback with the	574	Asynchronously unlink (delete) a file and call the callback with the
298	result code.	575	result code.
299		576
		577
		578	=item aio_mknod $path, $mode, $dev, $callback->($status)
		579
		580	[EXPERIMENTAL]
		581
		582	Asynchronously create a device node (or fifo). See mknod(2).
		583
		584	The only (POSIX-) portable way of calling this function is:
		585
		586	aio_mknod $path, IO::AIO::S_IFIFO | $mode, 0, sub { ...
		587
		588
300	=item aio_link $srcpath, $dstpath, $callback->($status)	589	=item aio_link $srcpath, $dstpath, $callback->($status)
301		590
302	Asynchronously create a new link to the existing object at C<$srcpath> at	591	Asynchronously create a new link to the existing object at C<$srcpath> at
303	the path C<$dstpath> and call the callback with the result code.	592	the path C<$dstpath> and call the callback with the result code.
304		593
		594
305	=item aio_symlink $srcpath, $dstpath, $callback->($status)	595	=item aio_symlink $srcpath, $dstpath, $callback->($status)
306		596
307	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at	597	Asynchronously create a new symbolic link to the existing object at C<$srcpath> at
308	the path C<$dstpath> and call the callback with the result code.	598	the path C<$dstpath> and call the callback with the result code.
309		599
		600
		601	=item aio_readlink $path, $callback->($link)
		602
		603	Asynchronously read the symlink specified by C<$path> and pass it to
		604	the callback. If an error occurs, nothing or undef gets passed to the
		605	callback.
		606
		607
310	=item aio_rename $srcpath, $dstpath, $callback->($status)	608	=item aio_rename $srcpath, $dstpath, $callback->($status)
311		609
312	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as	610	Asynchronously rename the object at C<$srcpath> to C<$dstpath>, just as
313	rename(2) and call the callback with the result code.	611	rename(2) and call the callback with the result code.
314		612
		613
		614	=item aio_mkdir $pathname, $mode, $callback->($status)
		615
		616	Asynchronously mkdir (create) a directory and call the callback with
		617	the result code. C<$mode> will be modified by the umask at the time the
		618	request is executed, so do not change your umask.
		619
		620
315	=item aio_rmdir $pathname, $callback->($status)	621	=item aio_rmdir $pathname, $callback->($status)
316		622
317	Asynchronously rmdir (delete) a directory and call the callback with the	623	Asynchronously rmdir (delete) a directory and call the callback with the
318	result code.	624	result code.
		625
319		626
320	=item aio_readdir $pathname, $callback->($entries)	627	=item aio_readdir $pathname, $callback->($entries)
321		628
322	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire	629	Unlike the POSIX call of the same name, C<aio_readdir> reads an entire
323	directory (i.e. opendir + readdir + closedir). The entries will not be	630	directory (i.e. opendir + readdir + closedir). The entries will not be
324	sorted, and will B<NOT> include the C<.> and C<..> entries.	631	sorted, and will B<NOT> include the C<.> and C<..> entries.
325		632
326	The callback a single argument which is either C<undef> or an array-ref	633	The callback is passed a single argument which is either C<undef> or an
327	with the filenames.	634	array-ref with the filenames.
		635
		636
		637	=item aio_readdirx $pathname, $flags, $callback->($entries, $flags)
		638
		639	Quite similar to C<aio_readdir>, but the C<$flags> argument allows to tune
		640	behaviour and output format. In case of an error, C<$entries> will be
		641	C<undef>.
		642
		643	The flags are a combination of the following constants, ORed together (the
		644	flags will also be passed to the callback, possibly modified):
		645
		646	=over 4
		647
		648	=item IO::AIO::READDIR_DENTS
		649
		650	When this flag is off, then the callback gets an arrayref with of names
		651	only (as with C<aio_readdir>), otherwise it gets an arrayref with
		652	C<[$name, $type, $inode]> arrayrefs, each describing a single directory
		653	entry in more detail.
		654
		655	C<$name> is the name of the entry.
		656
		657	C<$type> is one of the C<IO::AIO::DT_xxx> constants:
		658
		659	C<IO::AIO::DT_UNKNOWN>, C<IO::AIO::DT_FIFO>, C<IO::AIO::DT_CHR>, C<IO::AIO::DT_DIR>,
		660	C<IO::AIO::DT_BLK>, C<IO::AIO::DT_REG>, C<IO::AIO::DT_LNK>, C<IO::AIO::DT_SOCK>,
		661	C<IO::AIO::DT_WHT>.
		662
		663	C<IO::AIO::DT_UNKNOWN> means just that: readdir does not know. If you need to
		664	know, you have to run stat yourself. Also, for speed reasons, the C<$type>
		665	scalars are read-only: you can not modify them.
		666
		667	C<$inode> is the inode number (which might not be exact on systems with 64
		668	bit inode numbers and 32 bit perls). This field has unspecified content on
		669	systems that do not deliver the inode information.
		670
		671	=item IO::AIO::READDIR_DIRS_FIRST
		672
		673	When this flag is set, then the names will be returned in an order where
		674	likely directories come first. This is useful when you need to quickly
		675	find directories, or you want to find all directories while avoiding to
		676	stat() each entry.
		677
		678	If the system returns type information in readdir, then this is used
		679	to find directories directly. Otherwise, likely directories are files
		680	beginning with ".", or otherwise files with no dots, of which files with
		681	short names are tried first.
		682
		683	=item IO::AIO::READDIR_STAT_ORDER
		684
		685	When this flag is set, then the names will be returned in an order
		686	suitable for stat()'ing each one. That is, when you plan to stat()
		687	all files in the given directory, then the returned order will likely
		688	be fastest.
		689
		690	If both this flag and C<IO::AIO::READDIR_DIRS_FIRST> are specified, then
		691	the likely dirs come first, resulting in a less optimal stat order.
		692
		693	=item IO::AIO::READDIR_FOUND_UNKNOWN
		694
		695	This flag should not be set when calling C<aio_readdirx>. Instead, it
		696	is being set by C<aio_readdirx>, when any of the C<$type>'s found were
		697	C<IO::AIO::DT_UNKNOWN>. The absense of this flag therefore indicates that all
		698	C<$type>'s are known, which can be used to speed up some algorithms.
		699
		700	=back
		701
		702
		703	=item aio_load $path, $data, $callback->($status)
		704
		705	This is a composite request that tries to fully load the given file into
		706	memory. Status is the same as with aio_read.
		707
		708	=cut
		709
		710	sub aio_load($$;$) {
		711	my ($path, undef, $cb) = @_;
		712	my $data = \$_[1];
		713
		714	my $pri = aioreq_pri;
		715	my $grp = aio_group $cb;
		716
		717	aioreq_pri $pri;
		718	add $grp aio_open $path, O_RDONLY, 0, sub {
		719	my $fh = shift
		720	or return $grp->result (-1);
		721
		722	aioreq_pri $pri;
		723	add $grp aio_read $fh, 0, (-s $fh), $$data, 0, sub {
		724	$grp->result ($_[0]);
		725	};
		726	};
		727
		728	$grp
		729	}
		730
		731	=item aio_copy $srcpath, $dstpath, $callback->($status)
		732
		733	Try to copy the I<file> (directories not supported as either source or
		734	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		735	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		736
		737	This is a composite request that creates the destination file with
		738	mode 0200 and copies the contents of the source file into it using
		739	C<aio_sendfile>, followed by restoring atime, mtime, access mode and
		740	uid/gid, in that order.
		741
		742	If an error occurs, the partial destination file will be unlinked, if
		743	possible, except when setting atime, mtime, access mode and uid/gid, where
		744	errors are being ignored.
		745
		746	=cut
		747
		748	sub aio_copy($$;$) {
		749	my ($src, $dst, $cb) = @_;
		750
		751	my $pri = aioreq_pri;
		752	my $grp = aio_group $cb;
		753
		754	aioreq_pri $pri;
		755	add $grp aio_open $src, O_RDONLY, 0, sub {
		756	if (my $src_fh = $_[0]) {
		757	my @stat = stat $src_fh; # hmm, might block over nfs?
		758
		759	aioreq_pri $pri;
		760	add $grp aio_open $dst, O_CREAT | O_WRONLY | O_TRUNC, 0200, sub {
		761	if (my $dst_fh = $_[0]) {
		762	aioreq_pri $pri;
		763	add $grp aio_sendfile $dst_fh, $src_fh, 0, $stat[7], sub {
		764	if ($_[0] == $stat[7]) {
		765	$grp->result (0);
		766	close $src_fh;
		767
		768	my $ch = sub {
		769	aioreq_pri $pri;
		770	add $grp aio_chmod $dst_fh, $stat[2] & 07777, sub {
		771	aioreq_pri $pri;
		772	add $grp aio_chown $dst_fh, $stat[4], $stat[5], sub {
		773	aioreq_pri $pri;
		774	add $grp aio_close $dst_fh;
		775	}
		776	};
		777	};
		778
		779	aioreq_pri $pri;
		780	add $grp aio_utime $dst_fh, $stat[8], $stat[9], sub {
		781	if ($_[0] < 0 && $! == ENOSYS) {
		782	aioreq_pri $pri;
		783	add $grp aio_utime $dst, $stat[8], $stat[9], $ch;
		784	} else {
		785	$ch->();
		786	}
		787	};
		788	} else {
		789	$grp->result (-1);
		790	close $src_fh;
		791	close $dst_fh;
		792
		793	aioreq $pri;
		794	add $grp aio_unlink $dst;
		795	}
		796	};
		797	} else {
		798	$grp->result (-1);
		799	}
		800	},
		801
		802	} else {
		803	$grp->result (-1);
		804	}
		805	};
		806
		807	$grp
		808	}
		809
		810	=item aio_move $srcpath, $dstpath, $callback->($status)
		811
		812	Try to move the I<file> (directories not supported as either source or
		813	destination) from C<$srcpath> to C<$dstpath> and call the callback with
		814	a status of C<0> (ok) or C<-1> (error, see C<$!>).
		815
		816	This is a composite request that tries to rename(2) the file first; if
		817	rename fails with C<EXDEV>, it copies the file with C<aio_copy> and, if
		818	that is successful, unlinks the C<$srcpath>.
		819
		820	=cut
		821
		822	sub aio_move($$;$) {
		823	my ($src, $dst, $cb) = @_;
		824
		825	my $pri = aioreq_pri;
		826	my $grp = aio_group $cb;
		827
		828	aioreq_pri $pri;
		829	add $grp aio_rename $src, $dst, sub {
		830	if ($_[0] && $! == EXDEV) {
		831	aioreq_pri $pri;
		832	add $grp aio_copy $src, $dst, sub {
		833	$grp->result ($_[0]);
		834
		835	if (!$_[0]) {
		836	aioreq_pri $pri;
		837	add $grp aio_unlink $src;
		838	}
		839	};
		840	} else {
		841	$grp->result ($_[0]);
		842	}
		843	};
		844
		845	$grp
		846	}
328		847
329	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)	848	=item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs)
330		849
331	Scans a directory (similar to C<aio_readdir>) and tries to separate the	850	Scans a directory (similar to C<aio_readdir>) but additionally tries to
332	entries of directory C<$path> into two sets of names, ones you can recurse	851	efficiently separate the entries of directory C<$path> into two sets of
333	into (directories), and ones you cannot recurse into (everything else).	852	names, directories you can recurse into (directories), and ones you cannot
		853	recurse into (everything else, including symlinks to directories).
334		854
335	C<aio_scandir> is a composite request that consists of many	855	C<aio_scandir> is a composite request that creates of many sub requests_
336	aio-primitives. C<$maxreq> specifies the maximum number of outstanding	856	C<$maxreq> specifies the maximum number of outstanding aio requests that
337	aio requests that this function generates. If it is C<< <= 0 >>, then a	857	this function generates. If it is C<< <= 0 >>, then a suitable default
338	suitable default will be chosen (currently 8).	858	will be chosen (currently 4).
339		859
340	On error, the callback is called without arguments, otherwise it receives	860	On error, the callback is called without arguments, otherwise it receives
341	two array-refs with path-relative entry names.	861	two array-refs with path-relative entry names.
342		862
343	Example:	863	Example:
…		…
350		870
351	Implementation notes.	871	Implementation notes.
352		872
353	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.	873	The C<aio_readdir> cannot be avoided, but C<stat()>'ing every entry can.
354		874
		875	If readdir returns file type information, then this is used directly to
		876	find directories.
		877
355	After reading the directory, the modification time, size etc. of the	878	Otherwise, after reading the directory, the modification time, size etc.
356	directory before and after the readdir is checked, and if they match, the	879	of the directory before and after the readdir is checked, and if they
357	link count will be used to decide how many entries are directories (if	880	match (and isn't the current time), the link count will be used to decide
358	>= 2). Otherwise, no knowledge of the number of subdirectories will be	881	how many entries are directories (if >= 2). Otherwise, no knowledge of the
359	assumed.	882	number of subdirectories will be assumed.
360		883
361	Then entires will be sorted into likely directories (everything without a	884	Then entries will be sorted into likely directories a non-initial dot
362	non-initial dot) and likely non-directories (everything else). Then every	885	currently) and likely non-directories (see C<aio_readdirx>). Then every
363	entry + C</.> will be C<stat>'ed, likely directories first. This is often	886	entry plus an appended C</.> will be C<stat>'ed, likely directories first,
		887	in order of their inode numbers. If that succeeds, it assumes that the
		888	entry is a directory or a symlink to directory (which will be checked
		889	seperately). This is often faster than stat'ing the entry itself because
364	faster because filesystems might detect the type of the entry without	890	filesystems might detect the type of the entry without reading the inode
365	reading the inode data (e.g. ext2fs filetype feature). If that succeeds,	891	data (e.g. ext2fs filetype feature), even on systems that cannot return
366	it assumes that the entry is a directory or a symlink to directory (which	892	the filetype information on readdir.
367	will be checked seperately).
368		893
369	If the known number of directories has been reached, the rest of the	894	If the known number of directories (link count - 2) has been reached, the
370	entries is assumed to be non-directories.	895	rest of the entries is assumed to be non-directories.
		896
		897	This only works with certainty on POSIX (= UNIX) filesystems, which
		898	fortunately are the vast majority of filesystems around.
		899
		900	It will also likely work on non-POSIX filesystems with reduced efficiency
		901	as those tend to return 0 or 1 as link counts, which disables the
		902	directory counting heuristic.
371		903
372	=cut	904	=cut
373		905
374	sub aio_scandir($$$) {	906	sub aio_scandir($$;$) {
375	my ($path, $maxreq, $cb) = @_;	907	my ($path, $maxreq, $cb) = @_;
376		908
		909	my $pri = aioreq_pri;
		910
		911	my $grp = aio_group $cb;
		912
377	$maxreq = 8 if $maxreq <= 0;	913	$maxreq = 4 if $maxreq <= 0;
378		914
379	# stat once	915	# stat once
		916	aioreq_pri $pri;
380	aio_stat $path, sub {	917	add $grp aio_stat $path, sub {
381	return $cb->() if $_[0];	918	return $grp->result () if $_[0];
		919	my $now = time;
382	my $hash1 = join ":", (stat _)[0,1,3,7,9];	920	my $hash1 = join ":", (stat _)[0,1,3,7,9];
383		921
384	# read the directory entries	922	# read the directory entries
385	aio_readdir $path, sub {	923	aioreq_pri $pri;
		924	add $grp aio_readdirx $path, READDIR_DIRS_FIRST, sub {
386	my $entries = shift	925	my $entries = shift
387	or return $cb->();	926	or return $grp->result ();
388		927
389	# stat the dir another time	928	# stat the dir another time
		929	aioreq_pri $pri;
390	aio_stat $path, sub {	930	add $grp aio_stat $path, sub {
391	my $hash2 = join ":", (stat _)[0,1,3,7,9];	931	my $hash2 = join ":", (stat _)[0,1,3,7,9];
392		932
393	my $ndirs;	933	my $ndirs;
394		934
395	# take the slow route if anything looks fishy	935	# take the slow route if anything looks fishy
396	if ($hash1 ne $hash2) {	936	if ($hash1 ne $hash2 or (stat _)[9] == $now) {
397	$ndirs = -1;	937	$ndirs = -1;
398	} else {	938	} else {
399	# if nlink == 2, we are finished	939	# if nlink == 2, we are finished
400	# on non-posix-fs's, we rely on nlink < 2	940	# for non-posix-fs's, we rely on nlink < 2
401	$ndirs = (stat _)[3] - 2	941	$ndirs = (stat _)[3] - 2
402	or return $cb->([], $entries);	942	or return $grp->result ([], $entries);
403	}	943	}
404		944
405	# sort into likely dirs and likely nondirs
406	# dirs == files without ".", short entries first
407	$entries = [map $_->[0],
408	sort { $b->[1] cmp $a->[1] }
409	map [$_, sprintf "%s%04d", (/.\./ ? "1" : "0"), length],
410	@$entries];
411
412	my (@dirs, @nondirs);	945	my (@dirs, @nondirs);
413		946
414	my ($statcb, $schedcb);	947	my $statgrp = add $grp aio_group sub {
415	my $nreq = 0;	948	$grp->result (\@dirs, \@nondirs);
		949	};
416		950
417	$schedcb = sub {	951	limit $statgrp $maxreq;
418	if (@$entries) {	952	feed $statgrp sub {
419	if ($nreq < $maxreq) {	953	return unless @$entries;
420	my $ent = pop @$entries;	954	my $entry = shift @$entries;
		955
		956	aioreq_pri $pri;
		957	add $statgrp aio_stat "$path/$entry/.", sub {
		958	if ($_[0] < 0) {
		959	push @nondirs, $entry;
		960	} else {
		961	# need to check for real directory
		962	aioreq_pri $pri;
		963	add $statgrp aio_lstat "$path/$entry", sub {
		964	if (-d _) {
		965	push @dirs, $entry;
		966
		967	unless (--$ndirs) {
		968	push @nondirs, @$entries;
		969	feed $statgrp;
		970	}
		971	} else {
		972	push @nondirs, $entry;
		973	}
421	$nreq++;	974	}
422	aio_stat "$path/$ent/.", sub { $statcb->($_[0], $ent) };
423	}	975	}
424	} elsif (!$nreq) {
425	# finished
426	undef $statcb;
427	undef $schedcb;
428	$cb->(\@dirs, \@nondirs) if $cb;
429	undef $cb;
430	}	976	};
431	};	977	};
432	$statcb = sub {
433	my ($status, $entry) = @_;
434
435	if ($status < 0) {
436	$nreq--;
437	push @nondirs, $entry;
438	&$schedcb;
439	} else {
440	# need to check for real directory
441	aio_lstat "$path/$entry", sub {
442	$nreq--;
443
444	if (-d _) {
445	push @dirs, $entry;
446
447	if (!--$ndirs) {
448	push @nondirs, @$entries;
449	$entries = [];
450	}
451	} else {
452	push @nondirs, $entry;
453	}
454
455	&$schedcb;
456	}
457	}
458	};
459
460	&$schedcb while @$entries && $nreq < $maxreq;
461	};	978	};
462	};	979	};
463	};	980	};
		981
		982	$grp
464	}	983	}
		984
		985	=item aio_rmtree $path, $callback->($status)
		986
		987	Delete a directory tree starting (and including) C<$path>, return the
		988	status of the final C<rmdir> only. This is a composite request that
		989	uses C<aio_scandir> to recurse into and rmdir directories, and unlink
		990	everything else.
		991
		992	=cut
		993
		994	sub aio_rmtree;
		995	sub aio_rmtree($;$) {
		996	my ($path, $cb) = @_;
		997
		998	my $pri = aioreq_pri;
		999	my $grp = aio_group $cb;
		1000
		1001	aioreq_pri $pri;
		1002	add $grp aio_scandir $path, 0, sub {
		1003	my ($dirs, $nondirs) = @_;
		1004
		1005	my $dirgrp = aio_group sub {
		1006	add $grp aio_rmdir $path, sub {
		1007	$grp->result ($_[0]);
		1008	};
		1009	};
		1010
		1011	(aioreq_pri $pri), add $dirgrp aio_rmtree "$path/$_" for @$dirs;
		1012	(aioreq_pri $pri), add $dirgrp aio_unlink "$path/$_" for @$nondirs;
		1013
		1014	add $grp $dirgrp;
		1015	};
		1016
		1017	$grp
		1018	}
		1019
		1020	=item aio_sync $callback->($status)
		1021
		1022	Asynchronously call sync and call the callback when finished.
465		1023
466	=item aio_fsync $fh, $callback->($status)	1024	=item aio_fsync $fh, $callback->($status)
467		1025
468	Asynchronously call fsync on the given filehandle and call the callback	1026	Asynchronously call fsync on the given filehandle and call the callback
469	with the fsync result code.	1027	with the fsync result code.
…		…
474	callback with the fdatasync result code.	1032	callback with the fdatasync result code.
475		1033
476	If this call isn't available because your OS lacks it or it couldn't be	1034	If this call isn't available because your OS lacks it or it couldn't be
477	detected, it will be emulated by calling C<fsync> instead.	1035	detected, it will be emulated by calling C<fsync> instead.
478		1036
		1037	=item aio_sync_file_range $fh, $offset, $nbytes, $flags, $callback->($status)
		1038
		1039	Sync the data portion of the file specified by C<$offset> and C<$length>
		1040	to disk (but NOT the metadata), by calling the Linux-specific
		1041	sync_file_range call. If sync_file_range is not available or it returns
		1042	ENOSYS, then fdatasync or fsync is being substituted.
		1043
		1044	C<$flags> can be a combination of C<IO::AIO::SYNC_FILE_RANGE_WAIT_BEFORE>,
		1045	C<IO::AIO::SYNC_FILE_RANGE_WRITE> and
		1046	C<IO::AIO::SYNC_FILE_RANGE_WAIT_AFTER>: refer to the sync_file_range
		1047	manpage for details.
		1048
		1049	=item aio_pathsync $path, $callback->($status)
		1050
		1051	This request tries to open, fsync and close the given path. This is a
		1052	composite request intended to sync directories after directory operations
		1053	(E.g. rename). This might not work on all operating systems or have any
		1054	specific effect, but usually it makes sure that directory changes get
		1055	written to disc. It works for anything that can be opened for read-only,
		1056	not just directories.
		1057
		1058	Future versions of this function might fall back to other methods when
		1059	C<fsync> on the directory fails (such as calling C<sync>).
		1060
		1061	Passes C<0> when everything went ok, and C<-1> on error.
		1062
		1063	=cut
		1064
		1065	sub aio_pathsync($;$) {
		1066	my ($path, $cb) = @_;
		1067
		1068	my $pri = aioreq_pri;
		1069	my $grp = aio_group $cb;
		1070
		1071	aioreq_pri $pri;
		1072	add $grp aio_open $path, O_RDONLY, 0, sub {
		1073	my ($fh) = @_;
		1074	if ($fh) {
		1075	aioreq_pri $pri;
		1076	add $grp aio_fsync $fh, sub {
		1077	$grp->result ($_[0]);
		1078
		1079	aioreq_pri $pri;
		1080	add $grp aio_close $fh;
		1081	};
		1082	} else {
		1083	$grp->result (-1);
		1084	}
		1085	};
		1086
		1087	$grp
		1088	}
		1089
		1090	=item aio_msync $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1091
		1092	This is a rather advanced IO::AIO call, which only works on mmap(2)ed
		1093	scalars (see the C<IO::AIO::mmap> function, although it also works on data
		1094	scalars managed by the L<Sys::Mmap> or L<Mmap> modules, note that the
		1095	scalar must only be modified in-place while an aio operation is pending on
		1096	it).
		1097
		1098	It calls the C<msync> function of your OS, if available, with the memory
		1099	area starting at C<$offset> in the string and ending C<$length> bytes
		1100	later. If C<$length> is negative, counts from the end, and if C<$length>
		1101	is C<undef>, then it goes till the end of the string. The flags can be
		1102	a combination of C<IO::AIO::MS_ASYNC>, C<IO::AIO::MS_INVALIDATE> and
		1103	C<IO::AIO::MS_SYNC>.
		1104
		1105	=item aio_mtouch $scalar, $offset = 0, $length = undef, flags = 0, $callback->($status)
		1106
		1107	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1108	scalars.
		1109
		1110	It touches (reads or writes) all memory pages in the specified
		1111	range inside the scalar. All caveats and parameters are the same
		1112	as for C<aio_msync>, above, except for flags, which must be either
		1113	C<0> (which reads all pages and ensures they are instantiated) or
		1114	C<IO::AIO::MT_MODIFY>, which modifies the memory page s(by reading and
		1115	writing an octet from it, which dirties the page).
		1116
		1117	=item aio_mlock $scalar, $offset = 0, $length = undef, $callback->($status)
		1118
		1119	This is a rather advanced IO::AIO call, which works best on mmap(2)ed
		1120	scalars.
		1121
		1122	It reads in all the pages of the underlying storage into memory (if any)
		1123	and locks them, so they are not getting swapped/paged out or removed.
		1124
		1125	If C<$length> is undefined, then the scalar will be locked till the end.
		1126
		1127	On systems that do not implement C<mlock>, this function returns C<-1>
		1128	and sets errno to C<ENOSYS>.
		1129
		1130	Note that the corresponding C<munlock> is synchronous and is
		1131	documented under L<MISCELLANEOUS FUNCTIONS>.
		1132
		1133	Example: open a file, mmap and mlock it - both will be undone when
		1134	C<$data> gets destroyed.
		1135
		1136	open my $fh, "<", $path or die "$path: $!";
		1137	my $data;
		1138	IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
		1139	aio_mlock $data; # mlock in background
		1140
		1141	=item aio_mlockall $flags, $callback->($status)
		1142
		1143	Calls the C<mlockall> function with the given C<$flags> (a combination of
		1144	C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
		1145
		1146	On systems that do not implement C<mlockall>, this function returns C<-1>
		1147	and sets errno to C<ENOSYS>.
		1148
		1149	Note that the corresponding C<munlockall> is synchronous and is
		1150	documented under L<MISCELLANEOUS FUNCTIONS>.
		1151
		1152	Example: asynchronously lock all current and future pages into memory.
		1153
		1154	aio_mlockall IO::AIO::MCL_FUTURE;
		1155
		1156	=item aio_group $callback->(...)
		1157
		1158	This is a very special aio request: Instead of doing something, it is a
		1159	container for other aio requests, which is useful if you want to bundle
		1160	many requests into a single, composite, request with a definite callback
		1161	and the ability to cancel the whole request with its subrequests.
		1162
		1163	Returns an object of class L<IO::AIO::GRP>. See its documentation below
		1164	for more info.
		1165
		1166	Example:
		1167
		1168	my $grp = aio_group sub {
		1169	print "all stats done\n";
		1170	};
		1171
		1172	add $grp
		1173	(aio_stat ...),
		1174	(aio_stat ...),
		1175	...;
		1176
		1177	=item aio_nop $callback->()
		1178
		1179	This is a special request - it does nothing in itself and is only used for
		1180	side effects, such as when you want to add a dummy request to a group so
		1181	that finishing the requests in the group depends on executing the given
		1182	code.
		1183
		1184	While this request does nothing, it still goes through the execution
		1185	phase and still requires a worker thread. Thus, the callback will not
		1186	be executed immediately but only after other requests in the queue have
		1187	entered their execution phase. This can be used to measure request
		1188	latency.
		1189
		1190	=item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED*
		1191
		1192	Mainly used for debugging and benchmarking, this aio request puts one of
		1193	the request workers to sleep for the given time.
		1194
		1195	While it is theoretically handy to have simple I/O scheduling requests
		1196	like sleep and file handle readable/writable, the overhead this creates is
		1197	immense (it blocks a thread for a long time) so do not use this function
		1198	except to put your application under artificial I/O pressure.
		1199
479	=back	1200	=back
480		1201
		1202	=head2 IO::AIO::REQ CLASS
		1203
		1204	All non-aggregate C<aio_*> functions return an object of this class when
		1205	called in non-void context.
		1206
		1207	=over 4
		1208
		1209	=item cancel $req
		1210
		1211	Cancels the request, if possible. Has the effect of skipping execution
		1212	when entering the B<execute> state and skipping calling the callback when
		1213	entering the the B<result> state, but will leave the request otherwise
		1214	untouched (with the exception of readdir). That means that requests that
		1215	currently execute will not be stopped and resources held by the request
		1216	will not be freed prematurely.
		1217
		1218	=item cb $req $callback->(...)
		1219
		1220	Replace (or simply set) the callback registered to the request.
		1221
		1222	=back
		1223
		1224	=head2 IO::AIO::GRP CLASS
		1225
		1226	This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to
		1227	objects of this class, too.
		1228
		1229	A IO::AIO::GRP object is a special request that can contain multiple other
		1230	aio requests.
		1231
		1232	You create one by calling the C<aio_group> constructing function with a
		1233	callback that will be called when all contained requests have entered the
		1234	C<done> state:
		1235
		1236	my $grp = aio_group sub {
		1237	print "all requests are done\n";
		1238	};
		1239
		1240	You add requests by calling the C<add> method with one or more
		1241	C<IO::AIO::REQ> objects:
		1242
		1243	$grp->add (aio_unlink "...");
		1244
		1245	add $grp aio_stat "...", sub {
		1246	$_[0] or return $grp->result ("error");
		1247
		1248	# add another request dynamically, if first succeeded
		1249	add $grp aio_open "...", sub {
		1250	$grp->result ("ok");
		1251	};
		1252	};
		1253
		1254	This makes it very easy to create composite requests (see the source of
		1255	C<aio_move> for an application) that work and feel like simple requests.
		1256
		1257	=over 4
		1258
		1259	=item * The IO::AIO::GRP objects will be cleaned up during calls to
		1260	C<IO::AIO::poll_cb>, just like any other request.
		1261
		1262	=item * They can be canceled like any other request. Canceling will cancel not
		1263	only the request itself, but also all requests it contains.
		1264
		1265	=item * They can also can also be added to other IO::AIO::GRP objects.
		1266
		1267	=item * You must not add requests to a group from within the group callback (or
		1268	any later time).
		1269
		1270	=back
		1271
		1272	Their lifetime, simplified, looks like this: when they are empty, they
		1273	will finish very quickly. If they contain only requests that are in the
		1274	C<done> state, they will also finish. Otherwise they will continue to
		1275	exist.
		1276
		1277	That means after creating a group you have some time to add requests
		1278	(precisely before the callback has been invoked, which is only done within
		1279	the C<poll_cb>). And in the callbacks of those requests, you can add
		1280	further requests to the group. And only when all those requests have
		1281	finished will the the group itself finish.
		1282
		1283	=over 4
		1284
		1285	=item add $grp ...
		1286
		1287	=item $grp->add (...)
		1288
		1289	Add one or more requests to the group. Any type of L<IO::AIO::REQ> can
		1290	be added, including other groups, as long as you do not create circular
		1291	dependencies.
		1292
		1293	Returns all its arguments.
		1294
		1295	=item $grp->cancel_subs
		1296
		1297	Cancel all subrequests and clears any feeder, but not the group request
		1298	itself. Useful when you queued a lot of events but got a result early.
		1299
		1300	The group request will finish normally (you cannot add requests to the
		1301	group).
		1302
		1303	=item $grp->result (...)
		1304
		1305	Set the result value(s) that will be passed to the group callback when all
		1306	subrequests have finished and set the groups errno to the current value
		1307	of errno (just like calling C<errno> without an error number). By default,
		1308	no argument will be passed and errno is zero.
		1309
		1310	=item $grp->errno ([$errno])
		1311
		1312	Sets the group errno value to C<$errno>, or the current value of errno
		1313	when the argument is missing.
		1314
		1315	Every aio request has an associated errno value that is restored when
		1316	the callback is invoked. This method lets you change this value from its
		1317	default (0).
		1318
		1319	Calling C<result> will also set errno, so make sure you either set C<$!>
		1320	before the call to C<result>, or call c<errno> after it.
		1321
		1322	=item feed $grp $callback->($grp)
		1323
		1324	Sets a feeder/generator on this group: every group can have an attached
		1325	generator that generates requests if idle. The idea behind this is that,
		1326	although you could just queue as many requests as you want in a group,
		1327	this might starve other requests for a potentially long time. For example,
		1328	C<aio_scandir> might generate hundreds of thousands C<aio_stat> requests,
		1329	delaying any later requests for a long time.
		1330
		1331	To avoid this, and allow incremental generation of requests, you can
		1332	instead a group and set a feeder on it that generates those requests. The
		1333	feed callback will be called whenever there are few enough (see C<limit>,
		1334	below) requests active in the group itself and is expected to queue more
		1335	requests.
		1336
		1337	The feed callback can queue as many requests as it likes (i.e. C<add> does
		1338	not impose any limits).
		1339
		1340	If the feed does not queue more requests when called, it will be
		1341	automatically removed from the group.
		1342
		1343	If the feed limit is C<0> when this method is called, it will be set to
		1344	C<2> automatically.
		1345
		1346	Example:
		1347
		1348	# stat all files in @files, but only ever use four aio requests concurrently:
		1349
		1350	my $grp = aio_group sub { print "finished\n" };
		1351	limit $grp 4;
		1352	feed $grp sub {
		1353	my $file = pop @files
		1354	or return;
		1355
		1356	add $grp aio_stat $file, sub { ... };
		1357	};
		1358
		1359	=item limit $grp $num
		1360
		1361	Sets the feeder limit for the group: The feeder will be called whenever
		1362	the group contains less than this many requests.
		1363
		1364	Setting the limit to C<0> will pause the feeding process.
		1365
		1366	The default value for the limit is C<0>, but note that setting a feeder
		1367	automatically bumps it up to C<2>.
		1368
		1369	=back
		1370
481	=head2 SUPPORT FUNCTIONS	1371	=head2 SUPPORT FUNCTIONS
482		1372
		1373	=head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
		1374
483	=over 4	1375	=over 4
484		1376
485	=item $fileno = IO::AIO::poll_fileno	1377	=item $fileno = IO::AIO::poll_fileno
486		1378
487	Return the I<request result pipe file descriptor>. This filehandle must be	1379	Return the I<request result pipe file descriptor>. This filehandle must be
488	polled for reading by some mechanism outside this module (e.g. Event or	1380	polled for reading by some mechanism outside this module (e.g. EV, Glib,
489	select, see below or the SYNOPSIS). If the pipe becomes readable you have	1381	select and so on, see below or the SYNOPSIS). If the pipe becomes readable
490	to call C<poll_cb> to check the results.	1382	you have to call C<poll_cb> to check the results.
491		1383
492	See C<poll_cb> for an example.	1384	See C<poll_cb> for an example.
493		1385
494	=item IO::AIO::poll_cb	1386	=item IO::AIO::poll_cb
495		1387
496	Process all outstanding events on the result pipe. You have to call this	1388	Process some outstanding events on the result pipe. You have to call this
497	regularly. Returns the number of events processed. Returns immediately	1389	regularly. Returns C<0> if all events could be processed, or C<-1> if it
498	when no events are outstanding.	1390	returned earlier for whatever reason. Returns immediately when no events
		1391	are outstanding. The amount of events processed depends on the settings of
		1392	C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>.
		1393
		1394	If not all requests were processed for whatever reason, the filehandle
		1395	will still be ready when C<poll_cb> returns, so normally you don't have to
		1396	do anything special to have it called later.
499		1397
500	Example: Install an Event watcher that automatically calls	1398	Example: Install an Event watcher that automatically calls
501	IO::AIO::poll_cb with high priority:	1399	IO::AIO::poll_cb with high priority (more examples can be found in the
		1400	SYNOPSIS section, at the top of this document):
502		1401
503	Event->io (fd => IO::AIO::poll_fileno,	1402	Event->io (fd => IO::AIO::poll_fileno,
504	poll => 'r', async => 1,	1403	poll => 'r', async => 1,
505	cb => \&IO::AIO::poll_cb);	1404	cb => \&IO::AIO::poll_cb);
506		1405
507	=item IO::AIO::poll_wait	1406	=item IO::AIO::poll_wait
508		1407
		1408	If there are any outstanding requests and none of them in the result
509	Wait till the result filehandle becomes ready for reading (simply does a	1409	phase, wait till the result filehandle becomes ready for reading (simply
510	C<select> on the filehandle. This is useful if you want to synchronously wait	1410	does a C<select> on the filehandle. This is useful if you want to
511	for some requests to finish).	1411	synchronously wait for some requests to finish).
512		1412
513	See C<nreqs> for an example.	1413	See C<nreqs> for an example.
514		1414
		1415	=item IO::AIO::poll
		1416
		1417	Waits until some requests have been handled.
		1418
		1419	Returns the number of requests processed, but is otherwise strictly
		1420	equivalent to:
		1421
		1422	IO::AIO::poll_wait, IO::AIO::poll_cb
		1423
515	=item IO::AIO::nreqs	1424	=item IO::AIO::flush
516		1425
517	Returns the number of requests currently outstanding (i.e. for which their	1426	Wait till all outstanding AIO requests have been handled.
518	callback has not been invoked yet).
519		1427
520	Example: wait till there are no outstanding requests anymore:	1428	Strictly equivalent to:
521		1429
522	IO::AIO::poll_wait, IO::AIO::poll_cb	1430	IO::AIO::poll_wait, IO::AIO::poll_cb
523	while IO::AIO::nreqs;	1431	while IO::AIO::nreqs;
524		1432
525	=item IO::AIO::flush	1433	=item IO::AIO::max_poll_reqs $nreqs
526		1434
527	Wait till all outstanding AIO requests have been handled.	1435	=item IO::AIO::max_poll_time $seconds
528		1436
529	Strictly equivalent to:	1437	These set the maximum number of requests (default C<0>, meaning infinity)
		1438	that are being processed by C<IO::AIO::poll_cb> in one call, respectively
		1439	the maximum amount of time (default C<0>, meaning infinity) spent in
		1440	C<IO::AIO::poll_cb> to process requests (more correctly the mininum amount
		1441	of time C<poll_cb> is allowed to use).
530		1442
531	IO::AIO::poll_wait, IO::AIO::poll_cb	1443	Setting C<max_poll_time> to a non-zero value creates an overhead of one
532	while IO::AIO::nreqs;	1444	syscall per request processed, which is not normally a problem unless your
		1445	callbacks are really really fast or your OS is really really slow (I am
		1446	not mentioning Solaris here). Using C<max_poll_reqs> incurs no overhead.
533		1447
534	=item IO::AIO::poll	1448	Setting these is useful if you want to ensure some level of
		1449	interactiveness when perl is not fast enough to process all requests in
		1450	time.
535		1451
536	Waits until some requests have been handled.	1452	For interactive programs, values such as C<0.01> to C<0.1> should be fine.
537		1453
538	Strictly equivalent to:	1454	Example: Install an Event watcher that automatically calls
		1455	IO::AIO::poll_cb with low priority, to ensure that other parts of the
		1456	program get the CPU sometimes even under high AIO load.
539		1457
540	IO::AIO::poll_wait, IO::AIO::poll_cb	1458	# try not to spend much more than 0.1s in poll_cb
541	if IO::AIO::nreqs;	1459	IO::AIO::max_poll_time 0.1;
		1460
		1461	# use a low priority so other tasks have priority
		1462	Event->io (fd => IO::AIO::poll_fileno,
		1463	poll => 'r', nice => 1,
		1464	cb => &IO::AIO::poll_cb);
		1465
		1466	=back
		1467
		1468	=head3 CONTROLLING THE NUMBER OF THREADS
		1469
		1470	=over
542		1471
543	=item IO::AIO::min_parallel $nthreads	1472	=item IO::AIO::min_parallel $nthreads
544		1473
545	Set the minimum number of AIO threads to C<$nthreads>. The current default	1474	Set the minimum number of AIO threads to C<$nthreads>. The current
546	is C<4>, which means four asynchronous operations can be done at one time	1475	default is C<8>, which means eight asynchronous operations can execute
547	(the number of outstanding operations, however, is unlimited).	1476	concurrently at any one time (the number of outstanding requests,
		1477	however, is unlimited).
548		1478
549	IO::AIO starts threads only on demand, when an AIO request is queued and	1479	IO::AIO starts threads only on demand, when an AIO request is queued and
550	no free thread exists.	1480	no free thread exists. Please note that queueing up a hundred requests can
		1481	create demand for a hundred threads, even if it turns out that everything
		1482	is in the cache and could have been processed faster by a single thread.
551		1483
552	It is recommended to keep the number of threads low, as some Linux	1484	It is recommended to keep the number of threads relatively low, as some
553	kernel versions will scale negatively with the number of threads (higher	1485	Linux kernel versions will scale negatively with the number of threads
554	parallelity => MUCH higher latency). With current Linux 2.6 versions, 4-32	1486	(higher parallelity => MUCH higher latency). With current Linux 2.6
555	threads should be fine.	1487	versions, 4-32 threads should be fine.
556		1488
557	Under most circumstances you don't need to call this function, as the	1489	Under most circumstances you don't need to call this function, as the
558	module selects a default that is suitable for low to moderate load.	1490	module selects a default that is suitable for low to moderate load.
559		1491
560	=item IO::AIO::max_parallel $nthreads	1492	=item IO::AIO::max_parallel $nthreads
…		…
569	This module automatically runs C<max_parallel 0> at program end, to ensure	1501	This module automatically runs C<max_parallel 0> at program end, to ensure
570	that all threads are killed and that there are no outstanding requests.	1502	that all threads are killed and that there are no outstanding requests.
571		1503
572	Under normal circumstances you don't need to call this function.	1504	Under normal circumstances you don't need to call this function.
573		1505
		1506	=item IO::AIO::max_idle $nthreads
		1507
		1508	Limit the number of threads (default: 4) that are allowed to idle (i.e.,
		1509	threads that did not get a request to process within 10 seconds). That
		1510	means if a thread becomes idle while C<$nthreads> other threads are also
		1511	idle, it will free its resources and exit.
		1512
		1513	This is useful when you allow a large number of threads (e.g. 100 or 1000)
		1514	to allow for extremely high load situations, but want to free resources
		1515	under normal circumstances (1000 threads can easily consume 30MB of RAM).
		1516
		1517	The default is probably ok in most situations, especially if thread
		1518	creation is fast. If thread creation is very slow on your system you might
		1519	want to use larger values.
		1520
574	=item $oldnreqs = IO::AIO::max_outstanding $nreqs	1521	=item IO::AIO::max_outstanding $maxreqs
		1522
		1523	This is a very bad function to use in interactive programs because it
		1524	blocks, and a bad way to reduce concurrency because it is inexact: Better
		1525	use an C<aio_group> together with a feed callback.
575		1526
576	Sets the maximum number of outstanding requests to C<$nreqs>. If you	1527	Sets the maximum number of outstanding requests to C<$nreqs>. If you
577	try to queue up more than this number of requests, the caller will block until	1528	do queue up more than this number of requests, the next call to the
578	some requests have been handled.	1529	C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>)
		1530	function will block until the limit is no longer exceeded.
579		1531
580	The default is very large, so normally there is no practical limit. If you	1532	The default value is very large, so there is no practical limit on the
581	queue up many requests in a loop it often improves speed if you set	1533	number of outstanding requests.
582	this to a relatively low number, such as C<100>.
583		1534
584	Under normal circumstances you don't need to call this function.	1535	You can still queue as many requests as you want. Therefore,
		1536	C<max_outstanding> is mainly useful in simple scripts (with low values) or
		1537	as a stop gap to shield against fatal memory overflow (with large values).
585		1538
586	=back	1539	=back
587		1540
		1541	=head3 STATISTICAL INFORMATION
		1542
		1543	=over
		1544
		1545	=item IO::AIO::nreqs
		1546
		1547	Returns the number of requests currently in the ready, execute or pending
		1548	states (i.e. for which their callback has not been invoked yet).
		1549
		1550	Example: wait till there are no outstanding requests anymore:
		1551
		1552	IO::AIO::poll_wait, IO::AIO::poll_cb
		1553	while IO::AIO::nreqs;
		1554
		1555	=item IO::AIO::nready
		1556
		1557	Returns the number of requests currently in the ready state (not yet
		1558	executed).
		1559
		1560	=item IO::AIO::npending
		1561
		1562	Returns the number of requests currently in the pending state (executed,
		1563	but not yet processed by poll_cb).
		1564
		1565	=back
		1566
		1567	=head3 MISCELLANEOUS FUNCTIONS
		1568
		1569	IO::AIO implements some functions that might be useful, but are not
		1570	asynchronous.
		1571
		1572	=over 4
		1573
		1574	=item IO::AIO::sendfile $ofh, $ifh, $offset, $count
		1575
		1576	Calls the C<eio_sendfile_sync> function, which is like C<aio_sendfile>,
		1577	but is blocking (this makes most sense if you know the input data is
		1578	likely cached already and the output filehandle is set to non-blocking
		1579	operations).
		1580
		1581	Returns the number of bytes copied, or C<-1> on error.
		1582
		1583	=item IO::AIO::fadvise $fh, $offset, $len, $advice
		1584
		1585	Simply calls the C<posix_fadvise> function (see it's
		1586	manpage for details). The following advice constants are
		1587	avaiable: C<IO::AIO::FADV_NORMAL>, C<IO::AIO::FADV_SEQUENTIAL>,
		1588	C<IO::AIO::FADV_RANDOM>, C<IO::AIO::FADV_NOREUSE>,
		1589	C<IO::AIO::FADV_WILLNEED>, C<IO::AIO::FADV_DONTNEED>.
		1590
		1591	On systems that do not implement C<posix_fadvise>, this function returns
		1592	ENOSYS, otherwise the return value of C<posix_fadvise>.
		1593
		1594	=item IO::AIO::mmap $scalar, $length, $prot, $flags, $fh[, $offset]
		1595
		1596	Memory-maps a file (or anonymous memory range) and attaches it to the
		1597	given C<$scalar>, which will act like a string scalar.
		1598
		1599	The only operations allowed on the scalar are C<substr>/C<vec> that don't
		1600	change the string length, and most read-only operations such as copying it
		1601	or searching it with regexes and so on.
		1602
		1603	Anything else is unsafe and will, at best, result in memory leaks.
		1604
		1605	The memory map associated with the C<$scalar> is automatically removed
		1606	when the C<$scalar> is destroyed, or when the C<IO::AIO::mmap> or
		1607	C<IO::AIO::munmap> functions are called.
		1608
		1609	This calls the C<mmap>(2) function internally. See your system's manual
		1610	page for details on the C<$length>, C<$prot> and C<$flags> parameters.
		1611
		1612	The C<$length> must be larger than zero and smaller than the actual
		1613	filesize.
		1614
		1615	C<$prot> is a combination of C<IO::AIO::PROT_NONE>, C<IO::AIO::PROT_EXEC>,
		1616	C<IO::AIO::PROT_READ> and/or C<IO::AIO::PROT_WRITE>,
		1617
		1618	C<$flags> can be a combination of C<IO::AIO::MAP_SHARED> or
		1619	C<IO::AIO::MAP_PRIVATE>, or a number of system-specific flags (when
		1620	not available, the are defined as 0): C<IO::AIO::MAP_ANONYMOUS>
		1621	(which is set to C<MAP_ANON> if your system only provides this
		1622	constant), C<IO::AIO::MAP_HUGETLB>, C<IO::AIO::MAP_LOCKED>,
		1623	C<IO::AIO::MAP_NORESERVE>, C<IO::AIO::MAP_POPULATE> or
		1624	C<IO::AIO::MAP_NONBLOCK>
		1625
		1626	If C<$fh> is C<undef>, then a file descriptor of C<-1> is passed.
		1627
		1628	C<$offset> is the offset from the start of the file - it generally must be
		1629	a multiple of C<IO::AIO::PAGESIZE> and defaults to C<0>.
		1630
		1631	Example:
		1632
		1633	use Digest::MD5;
		1634	use IO::AIO;
		1635
		1636	open my $fh, "<verybigfile"
		1637	or die "$!";
		1638
		1639	IO::AIO::mmap my $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh
		1640	or die "verybigfile: $!";
		1641
		1642	my $fast_md5 = md5 $data;
		1643
		1644	=item IO::AIO::munmap $scalar
		1645
		1646	Removes a previous mmap and undefines the C<$scalar>.
		1647
		1648	=item IO::AIO::munlock $scalar, $offset = 0, $length = undef
		1649
		1650	Calls the C<munlock> function, undoing the effects of a previous
		1651	C<aio_mlock> call (see its description for details).
		1652
		1653	=item IO::AIO::munlockall
		1654
		1655	Calls the C<munlockall> function.
		1656
		1657	On systems that do not implement C<munlockall>, this function returns
		1658	ENOSYS, otherwise the return value of C<munlockall>.
		1659
		1660	=back
		1661
588	=cut	1662	=cut
589		1663
590	# support function to convert a fd into a perl filehandle
591	sub _fd2fh {
592	return undef if $_[0] < 0;
593
594	# try to generate nice filehandles
595	my $sym = "IO::AIO::fd#$_[0]";
596	local *$sym;
597
598	open *$sym, "+<&=$_[0]" # usually works under any unix
599	or open *$sym, "<&=$_[0]" # cygwin needs this
600	or open *$sym, ">&=$_[0]" # or this
601	or return undef;
602
603	*$sym
604	}
605
606	min_parallel 4;	1664	min_parallel 8;
607		1665
608	END {	1666	END { flush }
609	max_parallel 0;
610	}
611		1667
612	1;	1668	1;
613		1669
		1670	=head1 EVENT LOOP INTEGRATION
		1671
		1672	It is recommended to use L<AnyEvent::AIO> to integrate IO::AIO
		1673	automatically into many event loops:
		1674
		1675	# AnyEvent integration (EV, Event, Glib, Tk, POE, urxvt, pureperl...)
		1676	use AnyEvent::AIO;
		1677
		1678	You can also integrate IO::AIO manually into many event loops, here are
		1679	some examples of how to do this:
		1680
		1681	# EV integration
		1682	my $aio_w = EV::io IO::AIO::poll_fileno, EV::READ, \&IO::AIO::poll_cb;
		1683
		1684	# Event integration
		1685	Event->io (fd => IO::AIO::poll_fileno,
		1686	poll => 'r',
		1687	cb => \&IO::AIO::poll_cb);
		1688
		1689	# Glib/Gtk2 integration
		1690	add_watch Glib::IO IO::AIO::poll_fileno,
		1691	in => sub { IO::AIO::poll_cb; 1 };
		1692
		1693	# Tk integration
		1694	Tk::Event::IO->fileevent (IO::AIO::poll_fileno, "",
		1695	readable => \&IO::AIO::poll_cb);
		1696
		1697	# Danga::Socket integration
		1698	Danga::Socket->AddOtherFds (IO::AIO::poll_fileno =>
		1699	\&IO::AIO::poll_cb);
		1700
614	=head2 FORK BEHAVIOUR	1701	=head2 FORK BEHAVIOUR
		1702
		1703	This module should do "the right thing" when the process using it forks:
615		1704
616	Before the fork, IO::AIO enters a quiescent state where no requests	1705	Before the fork, IO::AIO enters a quiescent state where no requests
617	can be added in other threads and no results will be processed. After	1706	can be added in other threads and no results will be processed. After
618	the fork the parent simply leaves the quiescent state and continues	1707	the fork the parent simply leaves the quiescent state and continues
619	request/result processing, while the child clears the request/result	1708	request/result processing, while the child frees the request/result queue
620	queue (so the requests started before the fork will only be handled in	1709	(so that the requests started before the fork will only be handled in the
621	the parent). Threats will be started on demand until the limit ste in the	1710	parent). Threads will be started on demand until the limit set in the
622	parent process has been reached again.	1711	parent process has been reached again.
623		1712
		1713	In short: the parent will, after a short pause, continue as if fork had
		1714	not been called, while the child will act as if IO::AIO has not been used
		1715	yet.
		1716
		1717	=head2 MEMORY USAGE
		1718
		1719	Per-request usage:
		1720
		1721	Each aio request uses - depending on your architecture - around 100-200
		1722	bytes of memory. In addition, stat requests need a stat buffer (possibly
		1723	a few hundred bytes), readdir requires a result buffer and so on. Perl
		1724	scalars and other data passed into aio requests will also be locked and
		1725	will consume memory till the request has entered the done state.
		1726
		1727	This is not awfully much, so queuing lots of requests is not usually a
		1728	problem.
		1729
		1730	Per-thread usage:
		1731
		1732	In the execution phase, some aio requests require more memory for
		1733	temporary buffers, and each thread requires a stack and other data
		1734	structures (usually around 16k-128k, depending on the OS).
		1735
		1736	=head1 KNOWN BUGS
		1737
		1738	Known bugs will be fixed in the next release.
		1739
624	=head1 SEE ALSO	1740	=head1 SEE ALSO
625		1741
626	L<Coro>, L<Linux::AIO>.	1742	L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
		1743	more natural syntax.
627		1744
628	=head1 AUTHOR	1745	=head1 AUTHOR
629		1746
630	Marc Lehmann <schmorp@schmorp.de>	1747	Marc Lehmann <schmorp@schmorp.de>
631	http://home.schmorp.de/	1748	http://home.schmorp.de/

Diff Legend

–	Removed lines
+	Added lines
<	Changed lines
>	Changed lines